US 120933

I'll search for this patent number and related docket references.

Initial results returned mostly unrelated modern patents (e.g., US 9,073,933). Let me refine to the literal identifier and check the docket sources.

Let me run a few more targeted checks on the Google Patents record, inventor name, and any court/docket usage of this number.

Search outcome up front

I searched for the literal identifier 120933 across patent and court-oriented sources. What exists is US Patent 120,933 (US120933A) — an 1871 railroad patent. I found no CAFC 2026 docket, and no Federal Circuit or district-court document, keyed to the bare number "120933." I also found no USPTO record for this number other than the Google Patents full-text/facsimile record (which is the authoritative text reproduced in the prompt). Everything below is drawn from that record unless flagged.

Two disambiguation notes, stated literally per your rules:

• US120933D appearing in the Google Patents "Family Applications" table is the same document under a Google Patents internal document ID, not a separate patent ("D" suffix, same title, same date).
• Web results numbered "120933" that are not this patent: Seattle Council Bill 120933 (Ordinance 127191), Honolulu PD CID bulletin 120933, Santa Clara County item 120933, and modern patents whose numbers merely contain or resemble the digits (e.g., US 9,073,933; US 6,845,953; US 7,043,543). None of these are US 120,933.

Bibliographic data

Abstract

There is no abstract — US patents of this era were not published with one. The record's "Definitions" passage functions as the summary and reads in substance: the invention relates to a railway crossing, and its object is to make the crossing stronger and more durable than ordinary constructions in which the bases of the main rails and the guard-rails are simply bolted to the bed-plates, so that the bolts take all lateral strain; instead, the rail bases are seated in a groove or channel and the rail necks at the crossing junction are supported by a throat-piece or angle-iron.

Independent claims — plain language

Caveat on claim format: the specification as transcribed contains no formal numbered claim clause and no "What I claim is—" recital. Patents of this vintage often carried a single claim or none in the modern sense. The only claim-like statement is the "nature of this invention" paragraph, which I summarize below as the sole independent claim. I am explicitly flagging this rather than inventing claim numbers.

Sole statement of invention (the de facto claim): A railway crossing construction in which (a) the base or foot of each rail — including the rails at the crossing and the guard-rails — is fitted into a groove or channel formed in the face of the bed-plate, so that lateral strain on the rails is resisted by the sides of the grooves rather than by the fastening bolts; and (b) at the junction where the rails cross, the neck/web of the rails is braced by a throat-piece or angle-iron seated on the rail bases and shaped on its underside to fit the curvature of the rail neck at its junction with the base, so that part of the train's weight is borne by the throat-piece.

Plain-language breakdown of the disclosed features supporting that statement:

• Grooved bed-plate (C), iron, bolted to ties (A) by bolts (a): two parallel grooves (D, D′) with a narrow intermediate rib (a) forming one side of each groove, each groove receiving one rail base.
• Alternative single-groove variant: the rib may be dispensed with and the two grooves merged into one wide groove so the two rail bases lie close together and mutually support each other against lateral strain.
• Throat-piece/angle-iron (F) at the crossing junction, fitted between the rails and guard-rails of the two tracks, helping hold them in relative position and reinforcing the point where rail continuity is broken by intersection. Its underside is contoured to the rail neck curvature; a section of the rail base may alternatively be cut out so the throat-piece rests on the plate.
• Lateral-strain relief and fewer bolt holes: because groove walls take the side pressure, few bolts are needed (only to hold rails down), preserving the rail's original strength by avoiding many holes in the base.
• Flange clearance groove (b) cut through the middle of the face of the throat-piece so wheel flanges pass without interference.
• Renewability: a worn throat-piece can be replaced by removing the guard-rail and lifting it out.
• Stated scope note in the specification: "The crossing above described is that of a right angle, but the principle of the construction is [applicable otherwise]" — the sentence is truncated in the transcript.

Uncertainties I will not paper over

1. Filing date / application number: unknown. The authoritative record does not state a filing date and shows no application number. Do not infer one.
2. No assignee, no inventorship conflict found — but absence of an assignment record is not proof no assignment existed; it only means none is recorded in this source.
3. Trailing signature "D L. HUMPHREY" appears at the end of the transcribed text after the description. Its role (witness, attorney/agent, or OCR/transcription artifact) is not established by the record; I will not guess.
4. CAFC 2026 dockets: none found for "120933." A bare six-digit number is also not the format of a Federal Circuit docket number (those are typically YYYY-NNNN, e.g., 2026-1234), so a "CAFC 2026 docket 120933" would not be a valid citation as written.
5. The patent expired in 1888 and has been public domain for well over a century; there is no live prosecution, PTAB, or appeal posture that I can identify for this number.

Sources

• Google Patents full text and facsimile for US120933A, "Improvement in railway crossings," George W. Billings, Chicago, Ill., dated Nov. 14, 1871 (antedated Nov. 13, 1871): https://patents.google.com/patent/[US120933](/patent/US120933)/en
• Similar-documents listing on the same record includes other crossing/frog art: US 139,643 (1873, railroad crossings), US 218,632 (1879, railway-crossings), US 175,357 (1876, guard-rails for railroad-frogs).
• Searches for "120933" in court/PTAB-oriented material returned only unrelated matters (Seattle CB 120933/Ordinance 127191; Honolulu CID bulletin 120933; Santa Clara County item 120933) and unrelated modern patents, none of which reference US 120,933.

I'll search litigation databases and patent sources for this specific number.

Let me check litigation-specific databases and post-issuance proceedings more directly.

Let me do a few final targeted checks on the specific patent's litigation history and any historical suits.

Bottom line

No litigation is known — and none is likely to exist — involving U.S. Patent No. 120,933. I searched litigation-focused sources (Unified Patents Portal caselist, CourtListener/RECAP, Docket Alarm, Justia, and general patent-litigation indexes) and found zero district court, ITC, PTAB, or Federal Circuit proceedings naming this patent. I am not going to invent case names, docket numbers, or outcomes to fill that gap.

What the number literally is

Interpreting "120933" literally, without normalization, the matching record is:

The claimed subject matter is a railway crossing ("frog") in which the rail bases sit in grooves D, D' formed in the bed-plate, with an optional central rib a, plus a removable throat-piece/angle-iron F at the crossing junction of the rails.

Why no litigation exists

1. Terminal date. The patent was granted in 1871 and the statutory term ran out on or about November 14, 1888. Any infringement action would necessarily have been filed in that 17-year window, decades before PACER, CourtListener, Docket Alarm, or the Unified Patents Portal existed.
2. No post-issuance proceedings. The AIA-era PTAB (IPR/PGR/CBM) did not exist in 1888 and cannot reach a patent that expired in 1888. Nothing appears in the PTAB side of the Unified Patents Portal for this number.
3. Google Patents itself shows no litigation. The record's "Similar Documents" list is populated entirely with other 19th/early-20th century frog and rail patents (e.g., US 896,154; US 175,357; US 140,332; US 191,470; US 139,643; US 218,632; US 763,872), and there is no litigation field populated for US120933A.

Explicit false positives I discarded

The searches for "120933" surface many hits that are not US Patent 120,933 and must not be reported as litigation against it:

• US 6,945,933 ('933 patent) — Warsaw Orthopedic, Inc. v. NuVasive, Inc., S.D. Cal. No. 3:08-cv-01512, Fed. Cir. appeal involving spinal retractor patents. Different patent.
• U.S. Patent 9,173,054 and a " '933 Patent" — Microsoft v. Lenovo IPRs and parallel ITC/district court case concerning wireless communications. Different patent.
• Amgen " '933 patent" — Amgen v. Roche (D. Mass. 1:15-cv-13099 area) and Collegium oxycodone litigation. Different patent.
• WO 2009/120933 (Gwozdz) and WO 2014/120933 — PCT publications used as prior art in unrelated IPRs.
• Non-patent items numbered 120933: Seattle Council Bill 120933 (Ordinance 127191); a Honolulu PD CID bulletin; a Santa Clara County Board item; a Hungarian patent No. 120933 (low-pressure mercury vapor tube). Irrelevant.
• 2013 IL App (4th) 120933-U, R.L. Brink Corp. v. Grunloh — an Illinois state administrative case whose appellate docket number happens to end in 120933. Not a patent case.

One unverified lead (flagged, not asserted)

A mid-1800s railroad-frog appeal appears in volume 125 of U.S. Reports (https://www.supremecourt.gov/pdfs/USReports/USREPORTS-125_PDFA.pdf), discussing a frog with "two centre rails B B' joined to form the V-shaped point" united to wing rails by U-shaped channel irons. The excerpt does not identify the patent number, and I could not confirm it is US 120,933. The structure described (V-point rails plus channel irons) differs from Billings's grooved bed-plate and throat-piece F, so it more likely concerns a different patent. Do not treat this as litigation involving US 120,933.

Confidence and limits

• High confidence that no litigation is recorded for US 120,933 in modern databases: such a case could not appear in them, and none appears.
• Lower confidence regarding 19th-century court records. Pre-1900 federal patent suits were reported in Federal Cases and Off. Gaz., and full-text coverage of that era is incomplete. I did not find any reported suit on this patent, but I cannot rule out an unreported 1871–1888 circuit court action. A definitive negative would require a manual search of Federal Cases (West) and the Commissioner of Patents' Annual Reports for 1872–1889. If you want, I can attempt that specific archival search.

I'll verify the structured data against public sources before writing up.

Proceedings overview

Total AIA trial proceedings on file against US 120,933: zero. The structured "PTAB proceedings on file" block — sourced from the USPTO Open Data Portal and treated as canonical here — returns no Inter Partes Review, no Post-Grant Review, and no Covered Business Method review for this patent, in any status (no active, no claims-invalidated, no claims-sustained, no settled, no institution-denied). The breakdown is simply 0 / 0 / 0 / 0 / 0.

Bottom line for a defendant: there is no PTAB shield to raise and no PTAB finding to leverage — but there is also no patent left to be sued on. US 120,933 issued 1871-11-14, expired 1888-11-14 (17 years from issue, pre-1952 term), and has been public domain for roughly 138 years. Any demand letter citing "US 120,933" is not asserting a live patent; it is either a clerical/OCR error, a number-confusion scam, or a citation to a historical document. My search of PTAB-facing sources surfaced no proceeding keyed to US 120,933.

I want to be explicit about the two ways the requested output template does not fit the facts, so nothing here reads as a withheld or incomplete finding:

1. Chronological impossibility of AIA trials. The AIA trial regime (IPR under 35 U.S.C. §§ 311–319, PGR under §§ 321–329, CBM under AIA § 18) and the Board's trial rules at 37 C.F.R. Part 42 took effect 2012-09-16. This patent expired 1888-11-14. There was a 124-year gap in which no AIA trial vehicle existed, and the patent was already public domain for the entire life of the PTAB.
2. No modern claim set to try. As flagged in the previously generated claim section, the 1871 specification contains no numbered claims and no "What I claim is—" recital. An IPR petition must identify "each claim challenged" (37 C.F.R. § 42.104(b)) by number and request cancellation of claims. There are no claim numbers in this document to challenge, cancel, or sustain. Any FWD-style claim-level verdict ("claims 1–5 canceled") is structurally unavailable for this patent, not merely absent.

---

Proceedings on file (list)

None. There is no proceeding to report at any impact level. I am not padding this section with a "no activity" entry per proceeding, because inventing or inferring proceeding numbers is prohibited and the record supports only the null result.

For completeness, the following would each be a prerequisite for a reportable proceeding, and none is met: (a) a PTAB case number in the IPR20XX-NNNNN / PGR20XX-NNNNN / CBM20XX-NNNNN format — found: none for this patent; (b) an institution decision — found: none; (c) a Final Written Decision — found: none; (d) any Federal Circuit appeal of an FWD — found: none. I also note the format point already made in the prior section: a bare six-digit "120933" is not the format of a CAFC docket number (those are YYYY-NNNN), so a citation styled "CAFC 2026 docket 120933" is invalid as written regardless of subject matter.

Look-alike numbers with real PTAB activity (do not confuse these with US 120,933)

These are different documents. I list them only to head off misattribution, and I flag every detail below as search-result-level, not verified by me against the underlying Docket Alarm / PTAB E2E docket:

None of these references US 120,933 or George W. Billings, and I found no litigation, PTAB, or reexamination document that does.

---

Strategic summary

Claim status of US 120,933: there are no claims to categorize. Nothing is "CANCELED" by a Board panel, nothing is "SUSTAINED" against an invalidity challenge, and nothing is "UNTESTED." The document does not contain a modern claim set at all. The de facto sole statement of invention (the 1871 "nature of this invention" paragraph) stands as an unreviewed historical disclosure, not as an adjudicated or adjudicable claim. There is no surviving claim list to hand a defendant, because there was never a claim 1 to cancel.

Estoppel landscape: inapplicable. Section 315(e)(2) estoppel attaches only to a petitioner that obtained an FWD or that settled after institution. With no petitioner, no institution, and no FWD, there is no estoppel of any kind running against anyone. Symmetrically, there is no IPR-based invalidity finding binding on any party under § 315(e)(1). A defendant today has no § 315(e) bar to worry about because no IPR has ever been filed — and, as a practical matter, no need to raise prior art at all, since an 1888-expired patent cannot be enforced. The real "estoppel" risk for a defendant is not statutory; it is the risk of spending billable hours and a client's credibility treating a 1871 historical document as a live assertion.

Pattern signals: none of the flagged patterns are present. No petitioner has filed once, let alone multiple times. The patent owner (George W. Billings of Chicago, no recorded assignee) has pursued no PTAB appeals — he could not have, having died long before the PTAB existed. There is no defensive aggregator (Unified Patents, RPX, or similar) in the chain; such entities file IPRs against asserted patents, and this patent has been public domain since 1888, so there is nothing for them to challenge and no member exposure for them to abate. The absence of PTAB activity here is not the usual "no one has bothered yet" signal that might suggest an untested, still-hardened patent. It is the much stronger signal that the patent is not assertable and never has been in the AIA era.

---

Recommended next steps

• If a demand letter or complaint cites US 120,933: do not build an IPR strategy. Confirm the patent's stated term — the authoritative record shows issue 1871-11-14 and anticipated expiration 1888-11-14, i.e., "Expired – Lifetime" with a pre-1952 17-year term. There is no live right to infringe. Respond with the expiration date and, if the letter is part of a pattern, treat it as a candidate for a § 285 fee motion or a UDAP/state-law claim against the sender rather than a validity defense.
• If the number was mistranscribed or mis-OCR'd: check the likely intended target. The most plausible modern confusion candidate surfaced in search is US 12,015,933, which does have petition-stage IPR activity. Before doing anything substantive, verify the true patent number from the face of the letter/complaint and the issued patent, and re-run this analysis on the corrected number. Do not assume "120933" and "12,015,933" are the same document.
• No trial-stage milestones to track. Because no proceeding exists, there is no institution-decision deadline, no oral hearing date, and no statutory one-year FWD due date to calendar. Nothing is pending that could be stayed, joined, or terminated.
• Verify independently if you must. PTAB decisions and dockets are public at the Board's decision site (https://www.uspto.gov/patents/ptab/decisions) and the PTAB E2E / Patent Trial and Appeal Board End-to-End system (https://ptab.uspto.gov/); PTAB petitions are also posted in the PTAB petition repository at ptacts.uspto.gov. Federal Circuit dispositions are on the court's docket (https://cafc.uscourts.gov/) and on CourtListener (https://www.courtlistener.com/). A search of any of these for "120933" returns no proceeding concerning the 1871 Billings patent.
• Treat the null result as dispositive, not provisional. Per the prior section, the patent's "family" entry US120933D in the Google Patents record is the same document under an internal document ID, not a second patent with separate proceedings. The absence of PTAB activity here is a closed question, not an indexing gap.

---

Explicit uncertainties. (1) I could not verify any detail of the IPR2025-00464 Padagis v. Neurelis matter beyond search-snippet level; it is listed solely as a look-alike-citation disambiguation and is irrelevant to US 120,933. (2) The US 12,015,933 IPR referenced above is identified by a petition-stage caption in search results; I did not confirm its proceeding number, status, or disposition, and I am not asserting one. (3) Nothing in this section should be read as implying a proceeding number, panel, or FWD for US 120,933 that the record does not contain.

Sources consulted

• US120933A, "Improvement in railway crossings," George W. Billings, Chicago, Ill., issued 1871-11-14 (antedated 1871-11-13); "Expired – Lifetime," anticipated expiration 1888-11-14: https://patents.google.com/patent/[US120933](/patent/US120933)/en
• USPTO PTAB structured proceedings block (USPTO Open Data Portal), as supplied in this prompt — canonical and controlling: no AIA trial proceedings for US 120,933
• USPTO PTAB decisions portal: https://www.uspto.gov/patents/ptab/decisions
• PTAB E2E: https://ptab.uspto.gov/
• CourtListener: https://www.courtlistener.com/
• Search hits for the string "120933," all unrelated to US 120,933: US 12,015,933 IPR petition materials (ptacts.uspto.gov petitions/1558117); WO 2009/120933 (Gwozdz et al.) as EX1009 in Padagis US LLC v. Neurelis, Inc., IPR2025-00464 (https://www.docketalarm.com/cases/PTAB/IPR2025-00464/Padagis_US_LLC_v._Neurelis_Inc/)

I'll attempt to verify assignment records and any NPE/assertion history for this patent before writing.

Let me check for inventor background, any corporate context, and confirm the absence of assignment/legal-event records.

Let me verify the coverage window of the USPTO assignment database, since that materially affects whether absence of records is meaningful for an 1871 patent.

Inventors

Pattern checks and why they come back negative:

• Departure-within-12-months pattern (inventor leaves assignee, portfolio fire-sale follows): not applicable. This pattern requires a corporate assignee to depart from. There is none here.
• The only other name anywhere in the transcribed text is the trailing signature "D L. HUMPHREY." As flagged in the prior section, its role (witness, patent agent, or transcription artifact) is unresolved. I searched specifically for a Humphrey in connection with this patent and found nothing that would identify him. The closest analogue to a "correspondent of record" for an 1871 patent would be the prosecuting patent agent — and if Humphrey were that agent, he would be the only candidate correspondent for this chain. I cannot confirm that, and I will not assert it. No assignment document names him, because no assignment document exists in the searchable record.
• One piece of context that is not evidence but is worth recording: this patent issued five weeks after the Great Chicago Fire (October 8–10, 1871), which destroyed the Cook County records. That fire is a recognized reason pre-1872 Chicago commercial and notarial records are sparse. It does not affect the USPTO's own patent records for this patent, and I offer it only as background on why surviving local Chicago documentation for a plaintiff-side ownership chain would be thin. Nothing in the assignment record turns on it.

Original assignee

None. There is no original assignee.

The patent issued to George W. Billings as sole inventor/patentee, not to a company. This is consistent with the record: the Google Patents bibliographic panel shows no assignee field, no application number, and no assignment events.

Consequently, the customary sub-questions do not have answers in this record:

• Product embodying the claims — No product record exists. The specification describes a shop-fabricated iron bed-plate with machined grooves, a rib, and a renewable throat-piece. That is consistent with a foundry or railway-supply fabricator, but I found no evidence of any manufacturer, licensee, or vendor, and I will not infer one from the subject matter.
• Primary line of business — Undeterminable. Billings is identified only by residence.
• Current status of the assignee — No entity exists to have a status. The patent itself is Expired – Lifetime with an anticipated expiration of 1888-11-14 (the pre-1952 17-year term running from issue), i.e., 138 years in the public domain.

A related inventor-laterality note, carried forward rather than repeated: a different Billings (Albert M. Billings, also of Chicago) appears on a later, unrelated railway patent, US 248,840. That is a different inventor and is not a corporate successor, a name change, or an assignment link to US 120,933. I flag it only so nobody mistakes it for a chain link.

Assignment timeline

The USPTO Assignment Center has no recorded assignment for US 120,933 (US120933A). None. There is no reel/frame to report, no assignor, no assignee, no correspondent, and no execution or recordation date.

Stating this plainly, per the instruction to do so and stop:

• No assignment, security agreement, merger, change of name, license, or release is recorded against this patent in the searchable USPTO assignment corpus or in the Google Patents legal-events panel.
• The Google Patents legal-events list for US120933A contains only three entries — Application granted, Publication of US120933A, and Anticipated expiration (1888-11-14) — and no assignment event of any kind.

Two accuracy caveats I am required to state, because they cut in opposite directions:

1. Coverage window. The USPTO Patent Assignment Search / Assignment Center database contains recorded assignments from August 1980 to the present (the underlying bulk dataset is formally "Patent Assignment XML (1980 – Present)"). Per MPEP 301.01, pre-August-1980 assignment records are maintained in a separate historical database that is not served through the online search. This patent's entire conceivable assignment window (1871 → expiry in 1888) falls ~92 years before that coverage begins. So the online null result is, strictly speaking, uninformative about 1871–1888 transfers — it is a coverage artifact, not proof that Billings never assigned a territorial grant or an undivided interest.
2. The generic "no records means the original assignee still owns it" inference does not apply here. That heuristic assumes a live patent. This patent expired 1888-11-14 and has been unenforceable public-domain subject matter for well over a century. There is no owner to identify and no standing to establish.

I could not execute a live query against the Assignment Center UI. assignmentcenter.uspto.gov is a client-side JavaScript application, and my tooling for this task is web search plus the authoritative full text supplied in the prompt. I am therefore reporting what the indexed records show, not the output of a direct UI query. Verification link: https://assignmentcenter.uspto.gov/ (legacy mirror: https://assignment.uspto.gov/patent/index.html) — search by patent number 120933.

Timeline diagram

timeline
    title Ownership of US 120933
    1871 : Invented by George W Billings
         : Antedated Nov 13 1871
         : Patent issued Nov 14 1871
    1888 : Patent term expired Nov 14 1888
    1980 : USPTO online assignment records begin

Every dated entry above is drawn from the authoritative record; the 1980 entry is included solely to mark the boundary of the searchable assignment corpus, not as an event in this patent's chain.

NPE / troll-pattern signals

Because there is no recorded transfer of any kind, every signal below is assessed against an empty chain. I am stating each one explicitly rather than skipping, per the output spec.

Cross-checks performed and their results: Google Patents legal events (no assignment entries), USPTO PEDS/ODP (no prosecution posture to check — no application number is recorded), SEC EDGAR (no registrant exists in this chain), and the RPX / Unified Patents asserter directories (no hit for this patent number). The Google Patents "Similar Documents" list does include later crossing/frog art — US 139,643 (1873), US 218,632 (1879), US 175,357 (1876) — but that list is a technical-similarity citation list, not an ownership list. It carries no assignment or common-ownership data, and I am expressly not treating it as evidence of a portfolio, a family, or a shared owner.

Verdict

Insufficient data — no records, or only the original assignment.

Narrowly: the Patent Assignment Center holds zero recorded documents for US 120,933 — no reel, no frame, no assignor, no assignee, no correspondent — which is precisely the rubric's "no records" case, so the correct label among the five options is Insufficient data. That label should be read as a statement about the record, not a suggestion of any unresolved ownership question: this patent expired 1888-11-14, and because the searchable assignment corpus only begins in August 1980 (MPEP 301.01 relegates everything earlier to a separate historical database), even a genuine 1871–1888 transfer would not surface here. There is no live owner, no product, and no assertion history, and the honest conclusion is that the NPE analysis is inapplicable to a 138-year-expired patent rather than that a hidden chain is being concealed.

Verification: https://assignmentcenter.uspto.gov/ (legacy: https://assignment.uspto.gov/patent/index.html) — query patent number 120933; authoritative document at https://patents.google.com/patent/[US120933](/patent/US120933)/en.

I'll run targeted searches on the literal number and on the citation data for this record.

Let me probe the record's citation fields directly and hunt for pre-1871 crossing/frog art.

Prior Art Analysis — US Patent 120,933

Patent under analysis: US 120,933 A, "Improvement in railway crossings," George W. Billings, Chicago, Ill., granted November 14, 1871 (antedated November 13, 1871).

Cross-reference: the earlier sections of this analysis established that this patent carries no formal numbered claim and no "What I claim is—" recital; the sole claim-like statement is the "nature of this invention" paragraph. I do not repeat that analysis here, but it governs everything below — every §102 mapping is run against that single de facto claim statement, because there is nothing else to map against.

---

1. Search outcome for the literal number 120933 — the citation field is empty

I searched the number literally and inspected the fields of the record that mirrors the USPTO data.

Finding: US120933A contains no backward patent citations — none. Not one. The record at https://patents.google.com/patent/[US120933](/patent/US120933)/en has:

This is expected for a US patent of 1871. Google Patents' citation graph for pre-1900 US grants is essentially not populated, because citation data derives from examiner records and published front pages that did not exist in that form. So I cannot give you "each patent citation for 120933," because the answer is that there are none to give. I will not manufacture a citation list to fill the gap.

A note on the "Prior art keywords" field: those five words are Google Patents' own text-mined terms, not references cited by an examiner or applicant. Reporting them as "cited prior art" would be a category error.

Important consequence for your question: because there are no cited references, the only bona fide prior-art statement in the patent is the applicant's own admission of the prior construction practice. I treat that first, then reconstruct the closest contemporaneous art, filtered to pre-1871.

---

2. The patent's own admitted prior art (applicant admission, not a reference)

Source: specification of US 120,933, https://patents.google.com/patent/US120933/en

---

3. Reconstructed prior art — references that predate November 14, 1871

I ran down every pre-1871 item I could identify in the crossing/frog and rail-fastening art. For each: full citation, date, description, and the §102 paragraph(s) it would implicate.

3.1 US 61,243 — Sidney Parker (Chicago, Ill.), "Improved Railroad Frog"

3.2 US 76,365 — George Westinghouse, Jr., "Improved Railway-Frog" (reissued RE 3,564)

3.3 US 6,171 — Alfred B. Seymour (Bordentown, N.J.), "Improved combined rail-road bar"

3.4 GB patent A.D. 1866, No. 698 — William Thomson, "Improvements in the manufacture of railway crossings"

3.5 GB patent A.D. 1868, No. 47 — Myers & Cannot, rails secured in grooves in sleepers

3.6 GB patent A.D. 1868, No. 154 — Abel (for Poulet), crossings and switches

3.7 US 69,241 — "Improvement in joints of railroad-rails"

3.8 US 78,429 — Henry Chapin, "Railroad frog"

3.9 Scientific American item on frog/frog-plate construction

---

4. Items from the record's "Similar Documents" list that are NOT prior art here

Most of the similar documents are later than 1871 and therefore cannot be §102 art against US 120,933 (they could only be §102 art in the reverse direction, or §103-type context if this patent were being examined today, which it is not). Dates per the record:

US 139,643 (1873), US 140,332 (1873), US 175,357 (1876), US 191,470 (1877), US 211,193 (1879), US 218,632 (1879), US 229,522 (1880), US 304,702 (1884), US 323,430 (1885), US 349,978 (1886), US 462,399 (1891), US 511,784 (1894), US 522,867 (1894), US 573,543 (1896), US 620,696 (1899), US 763,872 (1904), US 779,617 (1905), US 795,737 (1905), US 840,697 (1907), US 896,154 (1908), US 926,133 (1909), US 990,199 (1911), US 1,017,315 (1912).

Only US 6,171 (1849) and US 69,241 (1867) from that list are pre-November 1871. Listing post-1871 art as "§102 prior art for US 120,933" would be a dating error, and I am declining to do it.

I am also not repeating here the modern-document false positives that earlier sections of this analysis already discarded (US 9,073,933; US 6,845,953; US 6,609,049; the IPR/ITC and district-court matters keyed to those; Seattle Council Bill 120933; Honolulu CID bulletin 120933; Santa Clara County item 120933; the Hungarian patent No. 120933; the Illinois appellate docket ending in 120933). Those cross-references stand as previously stated and none of them is prior art to US 120,933.

---

5. §102 mapping summary against the single de facto claim

Because the patent has no formal claim (per the earlier claim section), I map each reference against the two-element statement of invention: (A) rail bases seated in groove(s) in the bed-plate so that groove sides resist lateral strain, with few bolts; and (B) a throat-piece/angle-iron at the crossing junction seated on the rail bases, underside contoured to the rail-neck curvature, bearing part of the train's weight, removable by taking off the guard-rail, with a flange-clearance groove b through its face.

No single reference above anticipates the combined two-element statement. The closest single-reference challenge is US 61,243 (Parker) on element A, paired with US 76,365 (Westinghouse, Jr.) on the neck-support concept — but that is a §103-style combination argument, not §102 anticipation, and it is the only realistic attack I can construct from the art I was able to verify.

What remains outside all located art: (i) the specific recitation that the sides of the grooves resist "all the lateral strain," which is the stated novelty hook over the admitted bolted practice; (ii) the merge‑into‑one‑groove variant in which the two rail bases lie close together and "mutually support each other"; (iii) the throat-piece whose under-side is formed so as to fit the curvature of the neck of the rail in its junction with the base, expressly so that "the weight of the train upon the rails will be borne, in part, by the throat-piece"; and (iv) the renewability workflow — remove the guard-rail, lift out the worn throat-piece, drop in a new one. I found no pre-1871 reference disclosing (iii) or (iv).

---

6. Confidence and limits — stated plainly

• High confidence that US120933A contains zero cited references in the record. The citation fields are absent, not merely empty. This is a data-availability fact about pre-1900 US patents, not a judgment about the art.
• High confidence in the bibliographic dates of US 61,243 (Jan. 15, 1867), US 76,365 (Apr. 7, 1868), US 6,171 (1849), and the two British 1866/1868 abridgment entries, as retrieved.
• Moderate confidence in the GB 1866 No. 698 entry: the search-result snippet reads "1866, March 7.—No 698," and there is a second 1866 entry rendered as "March 19.—No 814" immediately after it in the same abridgment column. The juxtaposition is consistent with a March 1866 grant, but I have not seen the original printed abridgment page.
• Low confidence / unverified: US 69,241; US 78,429; US 61,158; the reissue number of Westinghouse US 76,365 (the OCR reads both "3,564" and "3,584"); and the substance of the Scientific American Dec. 28, 1867 frog item. These must not be cited as anticipating references without retrieving the underlying documents.
• Not established: the filing date and application number of US 120,933 are absent from the record (as flagged in the earlier section). This matters for §102(b): I cannot compute the one-year bar date. If the filing date were, say, mid‑1871, then art published before mid‑1870 would be §102(b) art and art published between mid‑1870 and the filing date would be §102(a) art. My §102(a)-vs-(b) labels above are therefore stated conditionally.
• Framing caveat: 35 U.S.C. § 102 did not exist in 1871; novelty then turned on R.S. § 4886 and the "known or used by others" / "patented or described in any printed publication" language. I have mapped to modern §102(a)/(b) and their AIA §102(a)(1)/(a)(2) counterparts by convention, as your question requests, and I flag that the mapping is conventional rather than historical.
• What I did not do: I did not return, and have not reported, results for similar numbers. Every reference above is either the exact patent under analysis, or a dated pre-1871 document I separately retrieved and identified by its own number.

Suggested next step if you want a defensible prior-art record: pull the full specifications of US 61,243 and US 76,365 (both facsimiles verified above), plus US 69,241 and US 78,429 (not yet read), and the original printed abridgment pages for GB 1866 No. 698 and GB 1868 Nos. 47 and 154. I can run that retrieval if you want it.

I'll ground this in the prior-art listing actually present on the record, then check which of those references could legally qualify as prior art to an 1871 patent.

Now let me pull the two references in the listing that actually predate the patent, plus check the state of pre-1871 frog art.

Obviousness analysis of US 120,933 under 35 U.S.C. § 103

Ground rules I have to state before the analysis

Three threshold points change how this question can be answered. I'm stating them up front rather than burying them, because two of them are substantive corrections to the framing of the task.

1. The "Prior Art" section of this record is not a prior-art citation list. On Google Patents, the block you are pointing me at is headed "Similar Documents." It is a machine-classifier output — documents Google's model judged textually/semantically similar to US120933A — not the references the 1871 examiner cited, and not a §102(b)/§102(a) prior-art chart. The record contains no "References Cited" or "Cited By" field populated with examiner citations. So the honest answer to "use the prior art section" is: I can use it, but I must first filter it for legal availability, and the filter removes almost all of it.

2. Twenty-three of the twenty-five listed "similar documents" postdate the patent and therefore cannot be prior art to it. Under §103 a reference must first qualify as prior art under §102 relative to the challenged patent's effective date. US 120,933 carries a dating caption of November 14, 1871, antedated November 13, 1871, and the record shows no filing date and no application number (this was flagged in the earlier bibliographic section and it matters here: with no filing date of record, the latest possible effective date is the 1871-11-13/14 date, which is the most defensive assumption for the patent owner, and I use it).

Only US 6,171 and US 69,241 are legally available. The rest are, at best, evidence of how the frog art subsequently developed — which is exactly the kind of hindsight evidence a §103 analysis must not use. (A post-dating reference can occasionally be used for what it shows about the level of ordinary skill, but that is a narrow doctrine and I would not build an obviousness case on 1900s frog patents against an 1871 patent.)

3. There are no "claims" to render obvious, and §103 is an anachronism for this document. Both points were established earlier in this analysis and I won't re-argue them. Practically: the 1871 specification has no numbered claims and no "What I claim is—" recital, so the analysis must run against the de facto claim — the "nature of this invention" paragraph — split into limitations. And the patent issued under the Act of July 8, 1870, roughly eighty-one years before §103 existed; the operative test in 1871 was the "ordinary mechanic" standard of Hotchkiss v. Greenwood, 52 U.S. (11 How.) 248 (1850), plus the 1870 Act's novelty/prior-art provisions. A modern §103 analysis is therefore a reconstruction of what a 19th-century court would have asked, mapped onto contemporary §103 vocabulary. I'll do it that way and say so.

---

The de facto claim, decomposed into limitations

From the specification's "nature of this invention" passage (the only claim-like statement in the document):

• L1 — A bed-plate (C), of iron, secured to ties/timbers (A) by bolts (a).
• L2 — The upper surface of the bed-plate formed with two parallel grooves D, D′ receiving the base/foot of the rails E.
• L3 — A narrow central rib a between the grooves, forming one side wall of each groove; the groove side walls, not the bolts, resist the lateral strain.
• L4 — Alternative: rib omitted, two grooves merged into one wide groove wide enough for both rail bases to lie close together and mutually support each other against lateral strain.
• L5 — A throat-piece or angle-iron F at the crossing junction, resting on the rail bases, its underside contoured to the curvature of the rail neck/web at its junction with the base, so the train's weight is partly borne by F and the rail neck is less liable to crush/buckle.
• L6 — Guard-rails held on the same bed-plate; the throat-piece fitted between the rails and guard-rails of the two tracks.
• L7 — A flange-clearance groove b cut through the middle of the face of the throat-piece.
• L8 — Renewability: worn throat-piece removable by lifting out the guard-rail.
• L9 — Fewer bolt holes → rail base strength preserved.
• L10 — Right-angle crossing, with a (truncated) statement that the principle extends to non-right-angle crossings.

---

Combination A — Seymour US 6,171 (1849) + the admitted prior-art crossing

This is the strongest available combination, because both halves of it are in the record or in the patent's own text.

What US 6,171 actually teaches (retrieved; quoted from the patent text at https://patents.google.com/patent/US6171):

• A rail formed in two halves divided on a vertical longitudinal plane, with "the face of one of the halves being formed with a fillet received and fitted in a corresponding groove in the other" — i.e., a groove-and-fillet mating pair used to stop relative lateral movement of rail parts under traffic.
• A groove formed at the junction of the two halves at the top so that "the metal when beaten down by the passage of trains may spread therein and avoid the straining of the securing-bolts by the spread of the two parts of the rail."
• An express statement of the problem: "Another fruitful source of destruction is to be found in the lateral action of the flanges of the wheels, which, acting on the separate sections, causes them to yield at their junctions."
• Express criticism of bolted chairs ("the ends of the sections where they are joined together by a chair rest on an iron bed, while the other portions rest on wood, and … the junction thus formed is very imperfect and yields to the action of trains").

What the patent itself admits as the starting art (this is admitted prior art under the specification-as-admission doctrine — an applicant's own description of "the ordinary way of constructing crossings of this class" is usable against the patent): "laying the rails down upon the bed-plate and thereto securing them by means of bolts passing through the base of the rails and bed-plate," with the stated objection that "all the lateral strain exerted upon the rails is borne by the bolts whereby said rails are fastened down to the plates," requiring "many bolts," whose "many holes bored in the base … weaken[] the strength of the rails."

Mapping:

Motivation to combine (the KSR "articulated reasoning"):

1. Same field of endeavor — both are railway permanent-way structures; both address the rail seat and the load path from wheel to foundation.
2. Same problem, expressly stated by both — Seymour names "lateral action of the flanges of the wheels" and "straining of the securing-bolts"; Billings names "the lateral strain exerted upon the rails … borne by the bolts." A person having ordinary skill in the art (a railway engineer/master mechanic of the 1860s) reading Seymour in 1871 would read it as directly addressing Billings's stated problem.
3. Predictable result, no change in principle of operation. Relieving fasteners by giving the rail foot a bearing shoulder against a wall is elementary statics — the thrust is taken in bearing/shear by the plate material instead of in tension/shear by the bolt shanks. Nothing in the combination changes how either device works.
4. Design incentive supplied by the art, not by hindsight — the trade-off "fewer holes in the rail base = stronger base" is not Billings's discovery; it is the premise of the very practice his specification criticizes.

Where Combination A fails. It gets L1–L4 and L9. It does not touch L5–L8 (the throat-piece/angle-iron and its flange groove) or L6 (guard-rail interaction). If the de facto claim is treated as a combination — which is how a 19th-century court would most likely have construed a patent of this description, and how the "combination vs. separate elements" line of authority ran — then Combination A alone does not render the whole de facto claim obvious. It renders obvious a bed-plate-and-groove crossing, which is the bulk of the disclosure.

---

Combination B — US 69,241 (1867) as the second reference

I do not know what US 69,241 discloses, and I will not guess. My search did not return its text; the record gives only the title "Improvement in joints of railroad-rails," the date 1867-09-24, and the citation. On title alone it is rail-joint art, not crossing/frog art, which would make it analogous art of the same field but aimed at a different structure (abutting rail ends rather than rail intersections). It is legally available (pre-1871), and a joint-bar/rail-joint patent from the late 1860s would be a natural place to look for teachings about (i) bearing a rail web against a shoulder, and (ii) avoiding bored holes in the base flange — both of which bear on L2/L3/L9. But that is inference from a title, not a disclosure analysis. To actually use this reference I would need its specification and drawing. I flag this as an open item rather than paper it over.

---

Combination C — the "merge the rib / single wide groove" variation (L3 vs. L4)

This one is obviousness on the patent's own terms, and it is the cleanest §103 point in the document.

The specification states: "Said rib, however, may be dispensed with and the two grooves thrown into one and made of sufficient width to admit the base of the two rails, so that they may lie close together, and thereby mutually support each other."

An inventor's own express statement that an element may be dispensed with and an alternative substituted is about as close to a concession of obviousness as a 19th- or 21st-century record gets. Removing a partition rib between two adjacent channels and making them one wider channel is a change in degree/form of the same structure performing the same function (admitting two rail bases side-by-side and resisting lateral thrust). The doctrine of the day — repeatedly invoked in this era and visible in the Federal Reporter material returned in my searches — was that a change in form, or the application of a known expedient to an analogous use, is not patentable invention; the excerpt from the Federal Reporter volume returned in my search states the principle that resolving a single combined tube back "into its original elements of two tubes" demanded "nothing more than … mechanical skill." So the rib/no-rib and one-groove/two-groove variants stand or fall together; if the two-groove-plus-rib version is obvious, the merged single groove is a fortiori obvious, and vice versa.

Caveat: this is a within-patent argument (obviousness of one disclosed variant over another disclosed variant), not an art-based §103 case. It is useful for showing that the L3/L4 distinction cannot carry patentable weight, but it does not dispose of L5.

---

Combination D — art for the throat-piece (L5–L8): the combination the record cannot support

Here is where the answer must be honest about a gap. The features that give US 120,933 whatever novelty it has are the throat-piece/angle-iron F — its underside contoured to the rail-neck curvature so it shares the wheel load — and its flange-clearance groove b, with the guard-rails of the two tracks holding it in place and allowing renewal.

Nothing in the legally available portion of the supplied prior-art section addresses any of that. The listed frog/crossing patents that would be the natural references for a throat-piece — US 140,332 (1873), US 139,643 (1873), US 175,357 (1876), US 191,470 (1877), US 211,193 (1879), US 218,632 (1879), US 349,978 (1886), US 511,784 (1894), US 763,872 (1904), US 840,697 (1907), US 896,154 (1908) — all postdate the patent and are legally unavailable. That is an unusual and instructive result: the classifier placed this patent squarely in the frog/crossing family, but the family's own earlier members are not in the list.

Supplemental art I located outside the supplied listing (labeled as such, because the task said to use the record's prior-art section and this is not it). These predate November 1871 and are relevant to L2/L3:

• US 32,794 (Smith, 1861) and US 36,579 (Smith, 1862) — cast-iron girders "bedded into the ground," with brackets and keys that confine the rail "both vertically and laterally," and with recesses formed in the upper surface of the girder for the rail/bracket seat, expressly to prevent "lateral or vertical displacement" without chair-bolts. This is close to L1–L3 in the permanent way (as opposed to rail-on-rail) locus.
• US 35,198 (Carpenter, 1862) — chairs whose jaws "are shaped in conformity with the section of the rails they are to hold," connecting rails "without the intervention of any bolts or keys," and expressly recognizing that bolted/keyed fastenings "so constantly" require renewal. Relevant to the motivation for L3/L9.
• US 20,793 (Hall, 1858) — a sole plate with lips for securing rail ends, i.e., a plate with raised formations that react against the rail foot.
• British abridgment, A.D. 1868, No. 47 (Myers & Cannot), Patents for Inventions: Abridgments of Specifications, Class Railways — "transverse sleepers … formed with grooves at a distance apart equal to the gauge of the railway … The rails are placed in the grooves and fixed in position therein by plates or clamps … as substitutes for chairs," the clamps having "ribs or flanges … formed as to enter and fit into the hollow on each side of the rail, and also in the cavity of the groove in the sleeper." This is a grooved-sleeper rail seat with flanking ribs — materially closer to L2/L3/L4 than anything in the US listing, and it is a printed publication predating 1871.

I retrieved these incidentally and have their relevant passages, but I did not verify them against examiner records or a formal prior-art search, and three of the four are outside the supplied section. Treat them as leads requiring verification, not as established grounds.

If the throat-piece art existed and were found, the motivation to combine is straightforward and non-hindsight: a frog's junction is the known point of maximum pounding and of broken rail continuity; adding a bearing block under the rail head/web at that point to share load is the obvious engineering response, and contouring it to the rail's underside is a routine fit-to-section step (Carpenter, 1862, already fits chair jaws to rail section). But I cannot identify the reference, so I cannot assert this. The honest statement is: the throat-piece limitation is the one limitation for which the supplied record provides no in-date anticipation or obviousness combination.

---

What the analysis actually yields

Net conclusion. If the de facto claim is construed broadly (any bed-plate with a channel receiving the rail foot so that the channel walls, rather than the bolts, take the side thrust), US 6,171 in view of the crossing practice the patent itself admits renders it obvious — and the "groove walls beat bolt shanks" mechanism is elementary enough that the combination requires no more than the skill of an ordinary mechanic under Hotchkiss. If the de facto claim is construed narrowly, as a combination including the throat-piece F — which is the construction a 19th-century court would likely have favored for a patent with no separately-claimed elements, and the construction that preserves validity — then the supplied prior-art section does not support an obviousness rejection of the whole, because the throat-piece is unaddressed by any in-date reference in that section. The validity of US 120,933 as an 1871 patent therefore turns on the throat-piece, not on the grooves.

Order of confidence: high on the L1–L4/L9 obviousness (grounded in retrieved text plus the patent's own admissions); low on any throat-piece rejection (no reference); high on the proposition that the 23 post-1871 "similar documents" cannot be combined against this patent.

---

Doctrinal and evidentiary cautions

1. No secondary considerations are in evidence. The specification asserts durability, reduced bolt-loosening, reduced rail-neck crushing, and renewability of the throat-piece, but asserts them without data. A 19th-century court weighing these would ask whether the asserted advantage is a new result or merely a better degree of a known result — the line of authority invoked, e.g., in the Federal Reporter excerpts returned in my searches (applying the principle that applying a known method to an analogous use, or changing form, is not invention). No long-felt-need, licensing, or commercial-success record exists in the sources I have. I found no such evidence; I am not asserting any.
2. Hindsight hazard. Because the natural frog art for the throat-piece is all post-1871, any throat-piece rejection would have to be built on art selected because it resembles F. That is precisely the hindsight the §103 framework forbids, and it is a further reason to leave that limitation as "not established" rather than to reason from the 1873–1908 frogs in the listing.
3. Effect of the missing filing date. If the actual 1871 filing/invention date were earlier than I have assumed, US 69,241 (1867-09-24) would still almost certainly qualify, and US 6,171 (1849) certainly would. The practical effect of the missing date on this analysis is small, but it is not zero, and I flagged it rather than inventing a date.
4. Use of this analysis. The patent expired 1888-11-14 and is public domain. This §103 work is therefore historical/validity-reconstruction, not a live invalidity defense; there is no claim set to cancel and no proceeding in which to raise it. That was established in the earlier sections and it constrains what anyone can do with the result.

---

Sources

• US120933A, "Improvement in railway crossings," George W. Billings, Chicago, Ill., dated 1871-11-14, antedated 1871-11-13; "Similar Documents" list and its dates; "Prior Art Keywords: rails, crossing, base, bolts, throat"; status "Expired – Lifetime," anticipated expiration 1888-11-14: https://patents.google.com/patent/[US120933](/patent/US120933)/en (authoritative for the patent text)
• US 6,171 A, Alfred B. Seymour, "Improved combined rail-road bar," 1849-03-13 — text retrieved and quoted: https://patents.google.com/patent/US6171
• US 69,241 A, "Improvement in joints of railroad-rails," 1867-09-24 — cited on the US120933A record; disclosure not retrieved; no content asserted
• US 32,794 (B. C. Smith, 1861-07-09) and US 36,579 (B. C. Smith, 1862-09-30) — girder/recess/bracket rail seats: https://patentimages.storage.googleapis.com/cf/92/82/29d99b12879430/US32794.pdf ; https://patentimages.storage.googleapis.com/28/dc/37/1c01e98aaab946/US36579.pdf
• US 35,198 (Hiram Carpenter, 1862-05-06) — chairs shaped to rail section, no bolts: https://patentimages.storage.googleapis.com/43/7f/62/b53ac4026c7d7c/US35198.pdf
• British abridgment, A.D. 1868, No. 47 (Myers & Cannot), Class Railways — grooved sleepers with clamping plates as chair substitutes: https://babel.hathitrust.org/cgi/ssd?id=uc1.c2540798;page=ssd;view=plaintext;seq=591
• Hotchkiss v. Greenwood, 52 U.S. (11 How.) 248 (1850) — "ordinary mechanic" standard, cited from general knowledge, not from the supplied record
• Federal Reporter excerpts returned in search (vol. 47 at 508 and 894; vol. 103 at 299; vol. 91 at 155) discussing analogous-use/form-change and "mechanical skill" reasoning — used only as illustrations of 19th-century doctrine; the specific cable-railway case quoted in one snippet is not identified and I do not attribute it

Explicit uncertainties. (a) I did not retrieve the disclosure of US 69,241 and have asserted nothing about it beyond its title and date. (b) The four supplemental pre-1871 references are outside the record's prior-art section and were located incidentally; their passages are quoted as retrieved, but they have not been verified as examiner-cited art and I do not treat them as established grounds. (c) No reference of any date was found that discloses a throat-piece/angle-iron contoured to the rail-neck curvature at a crossing with a wheel-flange clearance groove, so L5–L8 remain unrebutted in this analysis. (d) I have not re-verified the truncation in the specification's closing sentence about non-right-angle crossings.